Distributed Learning Algorithm for Interference Avoidance in Coexisting WBANs

Abstract

When multiple wireless body area networks (WBANs) exist in close proximity to each other, the inter-user interference considerably degrades the signal to interference plus noise ratio (SINR) of the hub nodes in the WBANs. One of the most effective coexistence schemes mentioned in the IEEE 802.15.6 WBAN standard is the channel hopping mechanism. However for a dense network of WBANs, the traditional random hopping mechanism specified in the standard doesn't consider its environment and thus fails to reduce the channel conflict among the coexisting WBANs. This exacerbates the inter-BAN interference when the operating frequencies are fewer. We propose a probabilistic channel hopping mechanism for a hub to select the channel with the least interference as the most probable for transmission. We formulate this channel selection problem as a finite repeated potential game and propose a distributed stateless Q-learning algorithm to achieve the Nash Equilibrium (NE). Numerical results show the convergence of the learning algorithm to the NE point of the game. Also, we find that the proposed algorithm significantly improves the network utility by reducing the interference when compared to the random hopping approach specified in the IEEE 802.15.6 standard.